Despite current aggressive treatment protocols based on irradiation and chemotherapy, the majority of women diagnosed with metastatic breast or ovarian cancer ultimately die. The objective of this proposal combines the technologies Of T cell immunobiology and synthetic peptide chemistry in a strategy of antigen discovery to identify individual candidate peptide sequences for ultimate use in the design of immunotherapeutic vaccines for metastatic breast and ovarian cancer. Individual clones of human cytotoxic T lymphocytes (CTL) having relevant lytic specificity for breast and/or ovarian tumor cells will be used to screen very large combinatorial peptide libraries for peptide sequences that best stimulate proliferative, cytokine release, or lytic responses by these CTL clones. Candidate sequences deduced from each clonal scan will be synthesized and each individually assessed for activity on the selecting clone. The most effective peptide sequences will then be made available for other necessary preclinical assessments and for phase I/II toxicity/efficacy clinical trials. Preliminary studies indicate that these library scans identify peptides several orders of magnitude more effective than native peptide ligands used to generate a particular T cell clone, and that these optimized superagonist peptides are effective as vaccines, provoking T cell mediated immune responses against native peptide ligands. PROPOSED COMMERCIAL APPLICATIONS: The survival statistics for women diagnosed with metastatic breast or ovarian cancer, following treatment with irradiation and chemotherapeutic drugs, are far from satisfactory. Alternative immunotherapeutic approaches have been hampered by the paucity of candidate tumor antigens, as well as the difficulties in identifying and optimizing them to enhance their effectiveness as vaccines. A strategy for antigen discovery is outlined that may substantially increase the availability of candidate tumor antigens for clinical tests as immunotherapeutic vaccines.